Adenosine has been reported to have cardioprotective (Olafsson et al., Circulation, 1987, 76:1135-1145) and neuroprotective properties (Dragunow and Faull, Trends in Pharmacol. Sci., 1988, 9:193; Marangos, Medical Hypothesis, 1990, 32:45). It is reportedly released from cells in response to alterations in the supply of or demand for oxygen (Schrader, Circulation, 1990, 81:389-391), is said to be a potent vasodilator, and is believed to be involved in the metabolic regulation of blood flow (Berne, Circ. Res., 1980, 47:808-813). However, adenosine has a short half life (&lt;1 sec) in human blood (Moser, et al., Am. J. Physiol., 1989, 256:C799-C806), and therefore high doses of adenosine would need to be administered continuously to achieve effective levels. Adenosine has been reported to exhibit negative inotropic, chronotropic and dromotropic effects (Belardinelli et al., Prog. in Cardiovasc. Diseases, 1989, 32:73-97) and to cause coronary steal by preferentially dilating vessels in nonischemic regions. Consequently, high doses of adenosine are toxic and this toxicity severely limits its therapeutic potential. However, it is believed that by increasing adenosine concentration locally, i.e. at the target site within the target tissue, the beneficial effects of adenosine can be provided without the toxic systemic effects. Adenosine has been reported to be an endogenous modulator of inflammation by virtue of its effects on stimulated granulocyte function (Cronstein etal., J. Clin. Invest., 1986, 78:760-770) and on macrophage, lymphocyte and platelet function. Adenosine receptor agonists have been reported to be beneficial in an experimental model of inflammation (Schrier, et al., J. Immunol., 1990, 145:1874-1879). Adenosine and a related analog have been reported to inhibit in vitro production of the cytokine, tumor necrosis factor alpha (Parmely et al., FASEB Journal, 1991, 5:A 1602).
Adenosine kinase is a cytosolic enzyme which catalyzes the phosphorylation of adenosine to AMP. Inhibition of adenosine kinase can potentially reduce the ability of the cell to utilize adenosine, leading to increased adenosine outside of the cell where it is pharmacologically active. However, the regulation of adenosine concentration is complex and involves other adenosine-metabolizing enzymes each with different kinetic properties and mechanisms of regulation. Adenosine can also be deaminated to inosine by adenosine deaminase (ADA) and condensed with L-homocysteine to S-adenosylhomocysteine (SAH) by SAH hydrolase. The role of each of these enzymes in modulating adenosine concentration is dependent on the prevailing physiological conditions, is tissue specific and is not well understood.
A number of nucleosides including purine, pyrrolo[2,3-d]pyrimidine and pyrazolo[3,4-d]pyrimidine analogs have been evaluated for inhibition of adenosine kinase but were reported to have K.sub.i 's of greater than 800 nM (Caldwell and Henderson, Cancer Chemother. Rep., 1971, 2:237-246; Miller et al., J. Biol. Chem., 1979, 254:2346-2352). A few compounds have been reported as potent inhibitors of adenosine kinase with Ki's of less than 100 nM. These are the purine nucleosides, 5'-amino-5'-deoxyadenosine (Miller et al., J. Biol. Chem., 1979, 254:2346-2352) and 1,12-bis(adenosin-N6-yl)dodecane (Prescott et al., Nucleosides & Nucleotides, 1989, 8:297), and the pyrrolopyrimidine nucleosides, 5-iodotubercidin (Henderson et al., Cancer Chemotherapy Rep. Part 2, 1972, 3:71-85; Bontemps et al., Proc. Natl. Acad. Sci. USA, 1983, 80:2829-2833; Davies et al., Biochem. Pharmacol., 1986, 35:3021-3029)and 5'-deoxy-5-iodotubercidin (Davies et al., Biochem. Pharmacol., 1984, 33:347-355; Davies et al., Biochem. Pharmacol., 1986, 35:3021-3029).
Some of these compounds have been used to evaluate whether adenosine kinase inhibition might lead to increased extracellular adenosine concentrations. In rat cardiomyocytes, inhibition of adenosine deaminase by 2'-deoxycoformycin was reported to have no effect on adenosine release from the cells. In contrast, inhibition of ADA together with adenosine kinase by 5'-amino-5'-deoxyadenosine resulted in a 6-fold increase in adenosine release (Zoref-Shani et al., J. Mol. Cell. Cardiol., 1988, 20:23-33). The effects of the adenosine kinase inhibitor alone were not reported. Similar results were reported in isolated guinea pig hearts; in these studies addition of 5'-amino-5'-deoxyadenosine to the perfusion medium, in the presence of EHNA to inhibit deamination, was reported to result in a 15-fold increase of adenosine release (Schrader, in Regulatory Function of Adenosine; (Berne et al.) eds. pp. 133-156, 1983). These effects were not apparent in the absence of ADA inhibition and other studies using isolated rat hearts perfused with 5-iodotubercidin alone, have reported no increase in perfusate adenosine concentration under normoxic conditions (Newby et al., Biochem. J., 1983, 214:317-323) or under hypoxic, anoxic or ischemic conditions (Achtenberg et al., Blochem, J., 1986, 235:13-17). In other studies, adenosine release has been measured in neuroblastoma cells in culture and compared with that of a variant deficient in adenosine kinase (AK-). The AK- cells used in this study were said to release adenosine at an accelerated rate; the concentration of adenosine in the growth medium was reported to be elevated compared to the normal cells (Green, J. Supramol. Structure, 1980, 13:175-182). In rat and guinea pig brain slices, adenosine uptake was reportedly inhibited by the adenosine kinase inhibitors, 5-iodotubercidin and 5'-deoxy-5-iodotubercidin (Davis et al., Biochem. Pharmacol., 1984, 33:347-355). However, inhibition of uptake and intracellular trapping via phosphorylation does not necessarily result in increased extracellular adenosine, since the adenosine could enter other metabolic pathways or the percentage of adenosine being phosphorylated could be insignificant compared to the total adenosine removed.
The effects of adenosine and certain inhibitors of adenosine catabolism, including 5-iodotubericidin were evaluated in an experimental model in which dog hearts were subjected to ischemia and reperfusion; 5-iodotubericidin was reported to have inconsistent effects (Wu, et al., Cytobios, 1987, 50:7-12).
Although the adenosine kinase inhibitors, 5'-amino-5'-deoxyadenosine and 5-iodotubercidin have been widely used in experimental models, the susceptibility of 5'-amino-5'-deoxyadenosine to deamination, and hence its potentially short half life, and the cytotoxicity of 5-iodotubercidin make their clinical utility limited and may limit interpretations based on these compounds. The pyrrolo[2,3-d]pyrimidines, 5-iodotubercidin and 5'-deoxy-5-iodotubercidin have been reported to cause pronounced general flaccidity and much-reduced spontaneous locomotor activity in mice, interpreted to be skeletal muscle relaxation; to cause hypothermia in mice; and to decrease blood pressure and heart rate in anesthetized rats (Daves et al., Biochem. Pharmacol., 1984, 33:347-355; Daves et al., Blochem. Pharmacol., 1986, 35:3021-3029; U.S. Pat. No. 4,455,420). The skeletal muscle effects of these compounds have been poorly documented, while the other effects were considered significant toxicities. It is believed that studies using these compounds were curtailed due to these toxicities and also because of their limited availability.